voice-extractor: add meta-instruction to SYNTHESIS_SYSTEM so generated outputs include "HOW TO APPLY" block

voice-extractor: document skill file installation methods in README

voice-extractor: fix contraction detection for possessives, document interactive --- delimiter

voice-extractor: fix paragraph detection, strengthen synthesis prompt from test failures

voice-extractor: add category profiles, preprocessing, PDF/DOCX support, and rule-generation improvements

Add voice-extractor tool and session rules to CLAUDE.md

Fix voice-cloner SKILL.md quality: clean excerpts, fix formatting, fix metrics

Add ignored workspace directory for voice-cloner samples and results

Add repo README with catalog of available agents and tools

Fix false positives in pre-commit-review flag detection

Add pre-commit-review agent for catching common issues before commit

Add engineering discipline rules to CLAUDE.md and SPEC.md

Add no-paid-API-calls rule to CLAUDE.md and SPEC.md

Add voice-cloner source files, eval corpus, and test suite

Replace LLM-based validation with deterministic LOO cross-validation